The regioselectivity of the ketal Claisen rearrangement - The Journal

G. William Daub, James P. Edwards, Carol R. Okada, Jana Westran Allen, Claudia Tata Maxey, Matthew S. Wells, Alexandra S. Goldstein, Michael J. Dibley...
0 downloads 0 Views 392KB Size
743

J. Org. Chem. 1982,47,743-145 10468-64-1; 2 (It6= CHB), 593-75-9;2 (R6 = t-Bu), 7188-38-7;38, 80242-76-8;3b, 80242-77-9;30, 80242-77-9;3d, 80242-79-1;3g, 80242-80-4;4b, 80242-81-5;5b, 80242-82-6;5d, 66175-28-8;6e, 80242-83-7; 6f, 80242-84-8; 7b, 80242-85-9; 8b, 80242-86-0.

Supplementary Material Available: Experimental details including IR and NMR spectral data and combustion analyses (5pages). Ordering information is given on any current masthead

Scheme I

$"

t

'"VR 5 ckR

MeOH (1)

t

2

1

Page.

Yoshihiko Ito, Hidehito Kato, Takeo Saegusa* Department of Synthetic Chemistry Faculty of Engineering Kyoto University Kyoto 606, Japan Received September 29, 1981

[3,31

cyR i

_c

Regioselectivity of the Ketal Claisen Rearrangement'

6

(1)Preaented in part at the 1981 Pacific Conference on Chemistry and Spectroscopy, Anaheim, Ca Oct 19-21,1981. (2)(a) Bennett, G. B. Synthesis 1977,589. (b) Zeigler, F. E. Acc. Chem. Res 1977,IO, 227. (3) Ireland, R. E.; Mueller, R. H.; Willard, A. K. J. Am. Chem. SOC. 1976,98,2868. (4) Johnson, W. S.; Werthemann, L.; Bartlett, W. R.; Brocksom, T. J.; Li, T.;Faulkner, D. J.; Peterson, M. R. J. Am., Chem. SOC.1970,92,741. (5)Felix, D.; Gexhwend-Steen, K.; Wick, A. E.; Eschenmoser, A. Helv. Chim. Acta 1969,52,1030. (6)(a) Werthemann, L.; Johnson, W. S. R o c . Natl. Acad. Sci. U.S.A. 1970,67, 1465,1810. (b) Loew, P.; Johnson, W. S. J. Am. Chem. SOC. 1971,93,3765. (7)Johnson, W. S.;Brocksom, T. J.; Loew, P.; Rich, D. H.; Werthemann, L.; Arnold, R. A.; Li, T.;Faulkner, D. J. J.Am. Chem. SOC.1970, 92,4463. (8)Faulkner, D. J.; Peterson, M. R. J. Am. Chem. SOC.1971,93,3766. (9)Johnson, W. 5.;Daub, G. W.; Lyle, T. A.; Niwa, M. J. Am. Chem. Soc. 1980,102,7800. (10)Saucy, G.;Marbet, R. Helu. Chim. Acta 1967,50,2091.

0022-3263/82/1947-0743$01.25/0

4'

3'

Scheme I1

Summary: The ketal Claisen rearrangement with a simple unsymmetrical ketal exhibits a high degree of regioselectivity, which is attenuated by substitution of the a-and &carbon atoms of the ketal.

Sir: The Claisen rearrangement has emerged as a very general and powerful synthetic tool over the last 10 years.2 In particular, enolate Claisen methods? ortho ester ketal exchange procedures,4 and amide acetal reactions have provided the synthetic chemist with convenient new methods for exploiting this historically important pathway to a,@-unsaturatedcarbonyl compounds. The ketal Claisen rearrangement has only been developed in a few specific cases. The work of Johnson and F a u l k n e P provide the only examples of the ketal Claisen rearrangement. The related enol-ether Claisen rearrangements from the work of Saucy'O are also included in this discussion because they involve nearly identical reaction pathways. For the more general case, the reaction between an acyclic unsymmetrical ketal (1) and an allylic alcohol (2) can give rise to two isomeric ketonic products. Scheme I details the mechanistic scenario for this process during which the intermediate cation i can be reversibly partitioned along two different pathways. These different paths lead to isomeric allyl/vinyl ethers (3 and 3') which irreversibly (Kw N 10s) rearrange to the isomeric ketones 4 and 4'. The ketal Claisen rearrangements developed by Johnson and Faulkner specifically avoid this problem, since one of the competing paths in each case is blocked."

(3)

[3,31

Ri

c

6

C6H5

C6H5

C6H5

7

5

example

a b

E

R,

R,

ketal

5% yield

Me t

H H H

di-Et

63

Et i-F'r t-Bu i-Bu Me

C

d e f g h

8

Et

di-Et

66

di-Me

84 74

H H H

di-Et

di-Me

di-Et

72 62

Me Me

di-Et di-Me

13

ratio 7:8 96:4 92:8 88:12 67:33 0:lOO 96:4 69:31 74:26

81

Scheme I11

I

1

9

C6H5

9'

10

C6H5

10'

Recent efforts in our laboratory have been designed to answer this regiochemical question, which is inherent in the ketal Claisen rearrangements of simple unsymmetrical ketals. Our preliminary work has examined the ketal Claisen rearrangements of some simple unsymmetrical ketals with (11)The three examples reported by Johnson and Faulkner (ref 6 8 ) lack hydrogens on one of the adjacent carbon atoms. The other example reported by Johnson (ref 9)effectively blocks the competing reaction path with a cyclopropyl group, which precludes the formation of an spacarbon atom at one of the a sites. Saucy's enol ether (ref 10) is symmetrical.

0 1982 American Chemical Society

744 J. Org. Chem., Vol. 47, No. 4,

Communications

Scheme IV

Scheme V H

H

c H3 C6H5

13 R = H or CH,

C6H5

14

cinnamyl alcohol. The ketal (6, 3.0 equiv) and cinnamyl alcohol (5) are heated to 120 "C in the presence of a catalytic amount (0.15 equiv) of propionic acid for 24 h. Subsequent workup and column chromatography12 on silica1 gel afford the isomeric reaction products. These products have been thoroughly ~haracterized.'~The results of this study are summarized in Scheme 11. The results in Scheme I1 present two significant trends. With the exception of example e, the more highly substituted (or "major") ketone (7) predominates over the less highly substituted (or "minor") one (8). The second trend is the decrease in regioselectivity as the a and 0carbons in the ketal (6) become more highly substituted. This effect is dramatically illustrated by the complete reversal of the selectivity in example e. These trends can be accomodated in a reasonably simple model. The intermediate cation (Scheme 111) is assumed to be in equilibrium with the isomeric allyl/vinyl ethers, which subsequently rearrange to the observed ketones. This assumption is quite reasonable, since one can show that the rate of protonation of the allyl/vinyl ethers is at least 2 orders of magnitude greater than the rate of the [3,3] sigmatropic rearrangement.I4 Consequently, there is a higher concentration of the more stable allyl/vinyl ether (9) and a greater proportion of the more highly substituted ketone (10). Additional substitution at the a or 0carbons lowers the selectivity as seen in Scheme 11. In the case of additional a substitution (examples g and h, Scheme 11), this is presumably due to the reduced rate of the [3,3] rearrangement for the more highly substituted allyl/vinyl ether. The work of Cressonls shows that substitution on the allyl terminus slows down the rate of sigmatropic rearrangement. Similarly, substitution of the vinylic terminus would be expected to retard the rearrangement step. This allows the less highly substituted path to compete more favorably. Hence the "minor" ketone 10' accounts for a larger proportion of the total product in examples g and h. Additional 0substitution develops unfavorable nonbonded interactions in the more highly substituted (12) The products were purified by medium-pressure liquid chromatography. (13) (a) All ketones exhibited 'NMR, IR, and low-resolution mass spectra that were consistent with their proposed structures. (b) Satisfactory combustion analyses were obtained for 7a, 7g, and 8g. (14) (a) The work of S a l ~ m a a "allows ~ for the approximation of the rate of protonation of the allyl/vinyl ether. The rate of protonation has a lower limit of 0.1 [allyl/vinyl ether] s-l at 120 OC. The work of Schmid'" allows for the approximation of the rate of sigmatropic rearrangement of the allyl/vinyl ether. This rate has an upper limit of 8.7 X lo-' [allyl/vinyl ether] s-*at 120 OC. (b) Kankaanperh, A.; Salomaa, P.; Juhala, P.; Aaltonen, R.; M a t u n , M. J . Am. Chem. SOC. 1973, 95, 3618. (c) Hansen, H. J.; Schmid, H. Tetrahedron 1974,30, 1959. (15) (a) Cresson, P.; Lemur, L. C. R.Acad. Sci., Ser. C 1966,262,1157. (b) Cresson, P.; Bancel, S. C. R. Acad. Sci., Ser. C 1968,266, 409.

E, Z

i H

1

CH3

erythro

allyl/vinyl ethers (11 and 12, Scheme IV). This results in a less favorable equilibrium constant for their formation. The alternative allyl/vinyl ether (13) is unaffected by the additional methyl groups and an increased proportion of the "minor" ketone (14) is observed. This effect is so strong that in case e the "minor" ketone can be isolated in a 72% yield. When the additional alkyl groups are placed on the y carbon (example f), the original selectivity is restored. The major product for all of the examples in Scheme I1 (except g) is obtained as a ca. 1:l mixture of the erythro and threo diastereomers. These are formed from the E,Z and E$ allyl/vinyl ethers, respectively, as shown in Scheme V. The work of Schmid'" suggests that the threo isomer should predominate due to the enhanced rate of [3,3] sigmatropic rearrangement of E,E isomers over E,Z isomers (kEE/kEZ= 3). However, Schmid measured these rates for allyl/vinyl ethers in which the vinyl group was disubstituted. In the current work the vinyl group is trisubstituted and the erythro/threo ratios for our exampled6 range from 1.2:l to 1:l. We have ruled out the possibility that acid-catalyzed equilibration of the threo and erythro isomers occurs during the reaction by subjecting a small amount of erythro-7f" to the reaction conditions. No threo-7f was detected by 'H NMR spectroscopy.'* These results demonstrate that ketal Claisen rearrangements with simple unsymmetrical ketals exhibit substantial selectivity and proceed in good yields. This selectivity may be as high as 251 in the most favorable cases. However, alkyl substitution on the a or 0 carbons of the ketal attenuate this selectivity and may even reverse it in unusual cases. We are currently investigating the complete scope of this reaction and its application to the synthesis of polyfunctional molecules.

Acknowledgment. Financial support for this work was kindly provided by the Research Corporation (Grant C1009 to G.W.D.), the National Science Foundation Undergraduate Research Participation Program (Grant to HMC Chemistry Department), and the Harvey Mudd College Chemistry Department. Registry No, 5, 104-54-1; 6 a , 52752-16-6; 6 b , 80359-80-4; 6c, 55904-98-8; 6 d , 80359-81-5; 6e, 72409-07-5; 6f, 80359-82-6; 6g, 80359-83-7; 6h,72409-06-4; erythro-7a, 32096-07-4; threo-7a, 3225297-4; erythro-7b, 80359-84-8; threo-7b, 80359-85-9; erythro-7d, (16)The erythro/threo ratios were determined by 'H NMR spectroscopy. The diastereomeric methyl ketone groups displayed distinctly different chemical shifta due to the strong anisotropic effect of the aromatic ring. The threo isomer exhibited a singlet at 8 2.10 for the methyl ketone protons, while the erythro isomer showed a higher field singlet at 6 1.85. (17) erythro-7f (99% pure by VPC) was obtained in selected fractions of the medium-pressure chromatography of 7f/8f. (18) The detection limits of 'H NMR spectroscopy for the presence of threo-7f were &2%.

J. Org. Chem. 1982,47,745-747 80359-86-0; threo-7d, 80359-87-1; erythro-7f, 80359-88-2; 7g, 8035989-3; erythro-7h, 80359-90-6; threo-7h, 80359-91-7; 88, 80359-92-8; 8b, 80359-93-9; 8d, 70245-09-9; 8e, 80359-94-0; 8f, 80359-95-1; 8g, 80359-96-2; 8h, 80359-97-3.

G . William Daub,* Michael G . Sanchez Robbin A. Cromer, Lester L. Gibson Department of Chemistry Harvey Mudd College Claremont, California 91711 Received October 27, 1981

745

initially formed ene adduct 2a undergoes a second, intramolecular ene reaction with the complexed aldehyde functioning as the enophile. Loss of methane from the resulting alcohol-Lewis acid complex to give the aluminum alkoxide prevents proton-catalyzed side reactions or solvolysis of the alcohol. Cyclization of 2a to 3a is much faster than formation of 2a since no 2a could be detected, even when the reaction is run to low conversion at -78 "C.

1

Sequential Ene Reactions. A New Annelation Procedure

Summary: Alkylidenecycloalkanes react with a,@-unsaturated carbonyl compounds in the presence of MezAICl to give bicyclic alcohols resulting from two sequential ene reactions.

Sir: The use of carbon-carbon double bonds as activating groups for the formation of new carbon-carbon bonds under mild conditions is a challenge to synthetic chemists. The ene reaction provides a potential solution to this problem.' We have found that Lewis acid catalyzed ene reactions with acrylate esters as the enophile occur at 25 "C and that the ene reactions of a-substituted acrylate esters are regioselective and stereoselective, with the carboalkoxy group adding endo.lb Lewis acid catalysis offers significant advantages over the corresponding thermal ene reactions which occur at 200-300 "C. We have also shown that alkylaluminum halides are preferred catalysts for these reactions since the alkyl group functions as a proton scavenger.2 a,@-Unsaturatedketones and aldehydes have seen very little use as en~philes.~Acrolein reacts with @-pineneat 140 "Csbor with ZnBr2 catalysis at 25 0C.4 Methyl vinyl ketone (MVK) reacts with @-pineneat 25 "C with ZnBr2 catalysis4 and has been reported to react with limonene with AlC1, ~atalysis.~ We report here that alkylidenecycloalkanes react with @-unsubstituteda,@-unsaturatedketones or aldehydes in the presence of MezAIClto give a bicyclic alcohol resulting from two sequential ene reactions. For instance, methylenecyclohexane (l),acrolein, and Me2AlCl in CHzClzat 0 "C for 20 min react to give a 63% yield of 3a.6vg The ~~

~

(1) (a) Hoffmann, H. M. R. Angew. Chem.,Znt. Ed. Engl. 1969,8,556. (b) Snider, B. B. Acc. Chem. Res. 1980,13,426. (2) Snider, B. B.; Rodini, D. J.; Karras, M.; Kirk, T. C.; Deutsch, E. A.; Cordova, R.; Price, R. T. Tetrahedron 1981,37, 3927. (3) (a) Albbtti, C. J.; Fisher, N. G.; Hogsed, M. J.; Joyce, R. M. J. Am. Chem. SOC. 1966, 78, 2637. (b) Kruk, C.; Velzen, J. C. v.; de Boer, T. J. Recl. Trau. Chim. Pays-Bas 1969.88, 139. (4) Snider, B. B. J. O g . Chem. 1974, 39, 255. (5) Mehta, G.; Reddy, A. V. Tetrahedron Lett. 1979,2625. We have been unable to repeat this reaction. Professor Mehta has indicated that "the yields are somewhat erratic and decrease enormously on scale up above 10 mmole." In addition the reported yield of 75% is baaed on recovery of 80% of the limonene. From the spectral data reported, there is no doubt that the ene adduct was obtained. However, our results make it clear that the ene adduct will not be stable to AlCl, for 12 h in benzene. The most likely explanation is that a partially hydrated, and therefore much less active, sample of AlCI, was used.

0022-3263/82/1947-0745$01.25/0

3

a, R, b, R, C, R, d, R, e, R, f, R, g, R,

= R, = H = H, R, = CH, = a-CH,, R, = H = 0-CH,, R, = H = 0-Br, R, = H = p-Br, R, = H = H, R, = CH,, ~ c Y - C H ,

Reaction of 1, MVK, and Me2AlClat -20 "C for 2 h gives a 39% yield of 2b and a 4% yield of 3b. The same reaction at 25 "C for 1h gives a 9% yield of 2b and a 49% yield of 3b. Thus, at different reaction times, either 2b or 3b can be isolated as the major product. The successful isolation of 2b, as opposed to 2a, results from the dimished reactivity of the ketone carbonyl as an enophile.'O The isolation of a teritary alcohol, 3b, from a Lewis acid catalyzed reaction is due to its protection as an aluminum alkoxide. Reaction of 1, methacrolein, and Me2AlCl (0 "C, 1 h) gives a 66% yield of a 3:l mixture of 3d (mp 69-71 "C) and 3c (mp 51.5-53.5 "C). The methyl group thus prefers to be equatorial, suggesting that ring formation is well ad(6) All new compounds were characterized by IR, 'H, and 13CNMR spectroscopy and gave satisfactory elemental analyses. The stereochemistry of hydroxyl and bromine substituents was established by the chemical shift and splitting pattern of the a-protons. The stereochemistry of these Substituents and methyl groups could be established by ISC NMR spectroscopy. The reported 13C spectra of 1,2,3,4,4a,5,6,'7-octacould be ashydronaphthalene and 1H-2,3,3a,4,5,6-he~ahydroindene~ signed using the spectra of 1-methylcyclohexene and 2-(methylmethylene)cyclohexane8as models. With shift values for equatorial and axial substituents on cyclohexanesand the shift values for equatorial and axial methyl substituents on methylenecyclohexane,8the expected 13C spectra could be calculated for each possible isomer. In all cases agreement between the calculated and observed spectra waa very good for all carbons. (7) Becker, K. B. Helu. Chim. Acta 1977, 60, 68. (8) Grover, S. H.; Stothers, J. B. Can. J. Chem. 1975,53, 589. (9) The detailed procedure given below is typical. Methylenecyclohexane (0.53 g, 5.5 mmol) was added to a solution of acrolein (distilled from CuSO,; 0.28 g, 5.0 mmol) and Me2AlCl (4.16 mL of 1.14 M in in 15 mL of CH2C12at 0 "C. The mixture was stirred heptane, 4.75 "01) for 20 min at 0 OC and quenched by cautious addition of water and ether. The organic layer waa removed and the aqueous layer was washed 3 times with ether. The combined organic layers were washed with brine, dried (Na2S04),and evaporated to give 0.808 g of crude product. Evaporative distillation of 0.696 g (100 O C , 0.25 torr) gave 0.411 g (63%) of pure 38: mp 54.4-55.5 OC; NMR (CCl,) 6 5.6 (br s, l), 3.8 (br 8, 1); 13C NMR 26.4,25.2, 21.9,20.9; I R (KBr) (CDCl,) 6 136.0, 124.3,71.3,42.1,34.8,33.5, 3370, 3050, 1670 cm-'. Anal. Calcd for ClnHIRO __ __ C, 78.90: H, 10.59. Found C, 78.79; H, 10.36. (10) Karras, M.; Snider, B. B. J. Am. Chem. SOC. 1980, 102, 7951.

0 1982 American Chemical Society